PRODUCE 4

Multivariate Analysis of Teachers’ Digital Information Skills - The

Importance of Available Resources

Roldan S. Cardona

MARY ROVELYN F. CABRALES, PHD ELE-SECTION 2

1. What is the research problem and what research design was

used by the researcher/s to address the said problem?

This study investigated the extent to which sociodemographic factors

and available resources predict digital information skills and which subgroups
emerge as potential targets for interventions, analyzing the interdependence
of the predictor variables.

Since one of the research gaps that has been highlighted in the study is
the limited availability of empirical research on teachers’ digital information
skills, it is justified to focus research interest specifically on this topic. Thus,
this study particularly intends to refine the image of teachers’ digital
information skills and to help target interventions on the most vulnerable
subgroups of teachers.

Given this intent, this study specifically, sought to answer the following
questions:

1. How do sociodemographic factors and available resources predict

teachers’ digital information skills?
2. What kind of teacher subgroups emerge as potential targets for
interventions by looking at how the interdependencies of multidimensional
factors divide teachers into distinct clusters?

Researchers in the field of education have been trying to give strong

emphasis on the need for multivariate approaches rather than mere bivariate
methods. In addition to this, it was even stressed by other researchers need
to consider the interdependencies of categorical predictors. To date, most of
the studies concerning teachers’ digital skills have relied only on self-report
data, recognizing the problem that people tend to misjudge their digital
competence, thus, the study has been pushed through to solve this existing
issue.

3. What are the research instruments used and how were they
developed and/or validated?

The study utilized a combination of self-reported survey responses and

performance test results to generate the data results of the study. This
complements existing research with a multivariate approach, combining
sociodemographic characteristics and the resources available to individuals,
generated by meso-level social activity. All these research instruments except
ICT Skill Test were validated through the use of Cronbach’s Alpha.

• ICT SKILL TEST

This tool was developed in the Research Unit for the Sociology of
Education which the researchers adopted it. This test collected the
teachers’ background information (age, gender, postal code of
place of employment, and information on whether the teacher is
a class or subject teacher).

• DIGITAL USAGE HABITS QUESTIONNAIRE

(Cronbach’s Alpha = .72).

This tool evaluated the frequency of the teachers’ ICT usage that
was rated on a scale of 0 = “never”, 1 = “sometimes, 2 =
“weekly”, 3 = “daily”, 4 = “several hours per day”.

• SELF-EFFICACY QUESTIONNAIRE
(Cronbach’s Alpha = .83)

This is the tool that measured whether or not the teachers

perceived themselves as competent in the areas of digital skills
(i.e., basic operations, office suite applications, communication,
networking and security, content creation and sharing, acquisition
and maintenance of applications and computational thinking)
related to their work.

• INSET TRAINING ADEQUACY QUESTIONNAIRE

(Cronbach’s Alpha = .86)
 This tool measured the adequacy of in-service training in the
above content areas in relation to the requirements of their
profession,

• PERFORMANCE-BASED TEST
(Cronbach’s Alpha = .78)

The performance-based test consists of 15 test items, each

producing a maximum of 2 points. Each item combines several
interactive, matching, multiple choice, and multiple-true-false
tasks, and thus none of the items is dichotomous in nature. This
study focuses on the 9 digital information skills related items of
the test.

They have provided the appendix A which presents the evidence

for the appropriate item difficulty and the sufficient discrimination
power. Likewise, the item-total correlation values indicate valid
internal consistency. Another evidence is also provided in the
same appendix that item characteristics are comparable across
gender, indicating proper measurement invariance.

4. How were the participants sampled or selected?

The data were collected in 2017–2019 in 549 schools from municipalities

belonging to two representative samples of Finnish municipalities formed by
the Finnish Education Evaluation Centre (FINEEC) based on municipal size and
regional location. The 2017–2018 sample consisted of 68 municipalities and
the 2019 sample of 69 municipalities, which in both samples is about a quarter
of Finnish municipalities.

Information about the study and participation links were provided to the
school leaders in the sample municipalities. Altogether 4988 teachers took
part in the study, which corresponds to a little over a tenth of Finnish basic
education teachers.

5. What multivariate statistics were used and why do you

think it is the most suitable statistics for the research problem?

The empirical data used in the study are taken from the questionnaire
responses and performance test results. Thus, the study employed MVA.

Multiple Regression Analysis

 At first, the suitability of multilevel modeling was tested. However, it
was found out that the variation of teachers’ ICT skills takes place almost
100% at the individual level, leaving no basis for such modeling. Thus,
multiple linear regression was chosen as the more appropriate approach in
relation to the available data.

In the process, all the variables were normalized by the min-max scalar
method to a scale of 0–1, and deviating values were removed from the
normalized variables, making them comparable and more suited for further
statistical analysis.

At first, a Pearson’s correlation table was created to examine the

descriptive statistics and preliminary interrelations of variables. In the
correlation table, the actual age was used instead of the normalized age.
Evaluation of the missing values showed that the missingness was
unsystematic and was therefore left untreated, leading to the exclusion of
those cases from the related analyses.

For me, the researcher chose the best MVA to treat the data because
the multiple linear regression allows the effect of multiple independent
variables on the variance of a continuous dependent variable to be examined.
The multiple regression analysis is an extension of simple linear regression.

The multiple regression analysis allows for relationship analysis, but

association within variables does not necessarily imply causation. (Nathans,
Oswald, & Nimon, 2012; Yan & Su, 2009). The multiple regression analysis
assumes that the observations are independent, the relationship between the
dependent variable and each independent variable is linear, the data shows
homoscedasticity but not multicollinearity, there are no significant outliers,
and the residuals (errors) are approximately normally distributed (Yan & Su,
2009).

6. What are the key findings of the research?

Through the MVA, the study found that digital activity and age explain most
of the variation in teachers’ digital information skills. Their effect was found to
be the inverse of each other, with abundant digital activity increasing and age
reducing teachers’ mastery of these skills.

Digital self-efficacy and in-service training also emerged as promising

predictors, highlighting that teachers’ information skills are explained more
strongly by the available resources than by sociodemographic factors. At the
end, the practical significance, and recommendations for more targeted
interventions as well as for further research are considered based on the
results obtained. The study found that teachers’ digital information skills are
explained more strongly by available resources than by sociodemographic
factors.

7. What are the strengths and weaknesses of the research

article?

In terms of both practical actions and future research, this research provides
a meaningful starting point, as education authorities can manage the
resources available, while sociodemographic factors as such inevitably fall
outside their sphere of influence. The digitalization of society and education
has a global potential to lead to an increase in the importance of the teaching
profession.

Introduction: The introduction is very comprehensive. It covered the

scope and the background of the study. This has shown the importance
and purpose of the study. The introduction emphasized why this
research matters in the context of education. If applicable, it is best that
the introduction should also connect to current issues in education to
show a problem with former theories or reveal a gap in current research.

Literature Review: A good academic paper should have a solid

literature review to bring about clarity and focus to the research
problem. This will in turn create a better research methodology and will
broaden the knowledge base in your research area. The study cited a
lot of related studies, but it has failed to establish coherence to make
sure one study links to another efficiently. This became a weak area that
can somehow affect the readers’ interest on the study.
Demographics:

The demographics show the totality of the respondents. and has

described the distribution of characteristics found. The study used
clustering to identify the different subgroups emphasizing the
interdependence and multidimensionality of influencing factors. The
research visualized different representations to understand the data set
and interpretations.

Data Collection Instruments:

One of the strengths of the paper is the tools they have employed to
carry out the research. Plus, the reliability test results for scales were
 tested and analyzed to make sure that the instrument used is reliable
to generate valid and reliable results. The overall research process,
models, and approaches are comprehensively discussed in the paper
which gives readers a clear understanding of what is going on in the
research and why it has arrived in that certain conclusion.

Statistical Tools:

The paper used a strong Multiple linear regression analysis performed

in this study to predict teachers’ information skills (dependent variable)
based on several independent variables, which were classified into two
groups: sociodemographic factors and available resources.

Contribution to the Body of Knowledge

Overall, it is a good, well-written research with an important message

for educational administrators, teachers, and learners in the field. The
piece, when taken as a whole, is relevant and very convincing in theory
and generated results that lay out a concrete way of approaching a problem
that has a lot of variables to compare. The researchers were able to
establish the purpose of the study and its significance to its contribution to
the body of knowledge that will help other teachers in the upskilling of their
ICT skills.
8. What learning on quantitative data analysis did you acquire from
the article?

Through the research study, I have seen how helpful multivariate

analysis is as a useful method to determine relationships and analyze
patterns among large sets of data as it takes a whole multitude of variables
into consideration. The study connects the variables accordingly with
appropriate methods. The biggest advantage of having such a model is that
it deems as many factors into consideration as possible. This results in a
great reduction of bias and provides accurate and specific results that are
close to reality. It is especially effective in reducing bias if a structured
study design is employed.

Because the study used a large data sample, it is good that it uses MVA to
efficiently do data reduction to ensure the simplification of the structure of
the research. In fact, the study was able to make simpler the data as much
as possible without sacrificing valuable information, which helps with the
interpretation of the findings.
 Moreover, since the study is dealing with multiple variables, it has perfectly
sorted and categorized similar ones together based on common
characteristics. Lastly, the study was able to investigate the variables
closely as they try to explore the data and better understand the
relationships between the variables of interest.

REFERENCE

Nathans, L. L., Oswald, F. L., & Nimon, K. (2012). Multiple linear regression:
A guidebook of variable importance. Practical Assessment, Research and
Evaluation, 17, 1–19. https://doi.org/10.7275/5fex-b874

Saikkonen, L., & Kaarakainen, M. T. (2021). Multivariate analysis of

teachers’ digital information skills - The importance of available
resources. Computers and Education, 168(March), 104206.
https://doi.org/10.1016/j.compedu.2021.104206

Yan, X. Su, X.(2009). Regression Analysis. Theory and Computing. New

Jersey.
https://books.google.com.ph/books?hl=en&lr=&id=5pdpDQAAQBAJ&oi=
fnd&pg=PR5&ots=CjEEhXczVm&sig=wqfBhbN2Y1a9-
S4yRseoGLB6vs4&redir_esc=y#v=onepage&q&f=false

Sir, kindly see the last meeting attendance, I failed to do it. I was not absent.
I reached out to Maam Pia she said, she will send me the link but failed to do
it. You may check the video recordings to double-check my attendance there.
Thank you very much.

Also, the 12% plagiarism result was due to the verbatim texts copy and
pasted from the research that is needed to answer the questions.
Interpretations and insights does not count on the plagiarism though. Thanks